• Why intentional? without intentionality in action, it's easy to get lost in the action and forget the purpose. Plus, when your intentional about experiencing new things, its easier to call yourself out on your bullshit ("I don't wanna try that cause I'm tired" -> "na it's cause I'm insecure") 
• Why non committal? if you commit to everything you experience, it's easy to get lost in that experience and forget why you were doing it in the first place. If you decide to experience a ton of drugs and forget to be non committal, you could become an addict.  

They fit into much larger buckets, this division is for granularity. Some of these are causal, ex: compassion leads to love, or fear leads to overcoming insecurity, or understanding leads to compassion. Logically there is strong causality, but there are nuances between them when experienced  emotionally. I find overcoming insecurity, hustle, euphoria, flow, status, love, and compassion the most fulfilling experiences based on scenarios I've lived in the past. My experience bucket is relatively dry though, I'm probably ignorant to true fulfillment.

We live and perceive in the 3rd dimension. When you think of a hypothetical population of people living in the second dimension, you think "how naive that they can't see volume". They can only see vertical segments and are unaware of the beauty that is space. But we're in that same situation. We're stuck in a single point in time: the present. We like to think we exist in the future and the past, but we don't. We're always in the present, and we can't escape it. Time is always passing us by and it's difficult for us to realize it cause we're derivatives on the larger graph. Imagine a 4rth dimensional society that could experience time and live in the past and present. They would find it so trivial that we can only the present and can't experience all of time as once. 

“Do I really look like a guy with a plan? I’m a dog chasing cars — I wouldn’t know what to do with one if I caught it”
- The Joker

I think it’s really easy for us to invent problems, then invent solutions to those imaginary problems. Here’s an example: instagram is great for sharing pictures with your friends and exploring what other people are sharing, but it’s difficult to find curated content. Why isn’t there an instagram specifically for animals? That way, all the pet owners could connect and create social circles with each other. There could be an “ask me anything” live stream feature where dog owners could ask experts on how to train their dogs. Plus, all the animal food / service companies would hop on and pay the most popular animals’ owners to promote their content. It makes sense right?

Not really. If you were really interested in animals, you could just follow specific pet own accounts on Instagram. Pet product promoters already exist, you wouldn’t be adding value to those companies. Plus, you can already reach out to people and create your own mini pet communities if that’s what you’re into, it’s not like instagram blocks all pet owners from coming in contact with each other.

“But it makes things so much easier… With Instapaw, you don’t need to go through the trouble of curating your own content and reaching out to communities of owners, we do that for you!”

That’s often the argument for any shitty idea. It “makes things easier”, “gives people more options”, “is curated for XYZ consumer instead”. In the end, it’s usually either an incremental improvement for a specific use case or just an over complicated feature for a market that isn’t interested.

Other times, we like to come up with ideas cause it’s fun. “Man imagine if I could do this! That would be so cool! I have to start building that right now”. We jump quickly into execution mode without being smart about how our time is being invested. I’ll give you an example: I built a CRM to track my meetings over the year the valuable information I learned from specific people. Seemed like a great idea, so I just started building. What I realize now is that a good indicator of a real problem is if you’ve actually tried a bunch of solutions before but they didn’t work. If I had tried spreadsheets, other tools like Dex, Notion, etc. and never found the right fit, then it would probably have been something worth pursuing. What I realized after I built it was that I was like a dog chasing cars, once I actually got to the car I didn’t know what to do. In the end, I didn’t end up really using my product. I was too excited by the process to focus on the outcome.

When it comes to figuring out if an idea is low quality or high potential, the best thing to do is to ask objective questions to figure out if there’s a market or not. “What is the problem?” for ex, shouldn’t hard to answer. There should be a very clear response to all of the questions below (stolen from Paul Graham, Sequoia, and Peter Thiel).

After building a few products that I never used, that didn’t work, and that didn’t have a proper market to gain traction, I thought I’d go one by one and figure out why each question is so important.

• Describe your company in 50 characters or less.

If it takes you too long to describe your company, usually the vision isn’t clear. “We’re a SaaS company that provides this service to allow for this thing but that also does this as a side effect to help that kind of person do this better…” When you can’t even properly explain your product in less than 50 characters, usually you don’t have a proper problem pinned down and are thus unclear on what your core value proposition is.

• What is the problem your company is trying to solve?

If you’re company isn’t solving some kind of problem, most likely people won’t go out of their way to use the product. It might be a cute “add-on” that gains some traction but won’t be a sustainable business model.

• Describe the pain of your customer — how is this addressed today and what are the shortcomings of current solutions?

A good indicator of if something is a problem or not is how much it emotionally affects people. If people spend 1hr a day on this thing, but it doesn’t bother them, then there’s less opportunity vs people spending 10min a day, but being super annoyed that they have to. Anger is always a good metric of how bad an experience is. If your core users are very angry with what they have to use today and you’re providing a better solution, than that’s usually a good sign.

• Why did you pick this idea? Do you have domain expertise in this area? How do you know people need what you’re making?

Snapchat is pretty easy to build. If you asked me to build it today, I could probably give you something pretty similar in 3–4 weeks. Same thing applies to instagram, twitter, facebook, etc. The reason they’re still around is because they pioneered the space of ___ (social networking, blogging with pictures, chatting with images, etc.), but if they went out of business tomorrow, you would see hundreds of other companies reabsorbing their users. The only reason people aren’t switching away from facebook today is because they already have a bunch of friends and communities on the platform and couldn’t leave unless everybody else was on another platform.

If you have very specific domain expertise in the field where you’ve found an opportunity, you’ll have a competitive advantage over anyone else trying to solve the same problem. You’ll also have a much better grasp on the problem and how to fix it since you’ve probably experienced it first hand.

• What do you understand about your business that other companies in it just don’t get?

Goes hand in hand with the last question. What domaine expertise do you have on this industry that your competitors don’t? What places you at an advantage vs competitors? What makes you capable of scaling this product?

• What’s new about what you’re making? What substitutes do people resort to because it doesn’t exist yet?

If you’re re-creating a product that already exists, you’re chances of success are much lower. 1) If the company has been around for a while and has a solid team but doesn’t have many users, then they probably just don’t have product market fit in which case you should pivot away from that idea. 2) If the company exists and has had previous growth but isn’t growing any more, then they’ve probably reached the majority of users in that niche market and you need to figure out a better way to solve the problem that expands the market. 3) If the company is pretty young and has had some pretty good success already (user retention, revenue, NPS, etc) and isn’t showing signs of slowing down, then you might want to be the competitor if you can execute better and have a higher chance of bringing your vision to fruition.

But as a general rule, you want to avoid recreating a product that already exists. You need to have some sort of unique form of value add to the user. If users are using shitty products as substitutes, like Excel or Drive, and are frustrated, then you’re probably on to something.

• Why now? Nature hates a vacuum — so why hasn’t your solution been built before now?

Why is today the best time to pursue this idea? Why couldn’t it have been done before? Why couldn’t it be done in the future? Maybe before code wasn’t open source, or there were no APIs for XYZ, or whatever it might be. If the product is easy to build and has never been attempted in the past, usually thats a bad sign. Low hanging fruits are collected 99% of the time. If it could have been built before but wasn’t, it’s probably because someone tried and failed. In which case, maybe they sucked at executing, scaling the business, recruiting talent, or whatever it might be and that’s why they failed. But usually, someone with the right skill set would have identified the problem and stepped in.

• How will you get users, and if you’re a marketplace how will you overcome the chicken and egg problem?

This has been a big road block for me in the past. For example, I built a platform for daily updates, basically a way to keep yourself and your close friends accountable on their progress updates. I was the target user, since I always hated sending daily updates on slack and other mediums. I built a bot that would automate sending for me, but then I was thinking “why doesn’t everyone use a bot like this?”, so I just built a centralized platform. Problem is, since it’s a very social app, people won’t use it unless others do, and it’s quite a pain to convince all your friends to move to another platform, so may as well just stay on slack. Paypal had a similar chicken and egg problem where they wanted people on ebay and craigslist to use paypal as their form of payment, but no one would sign up unless the person they were selling to or buying from signed up. To hack this, they bought a bunch of random small stuff on ebay and craigslist and asked the sellors to get paypal so they could make the exchange. That created a domino effect that kept on growing as people would ask other people on the platform to use paypal too.

• Are you starting with a big share of a small market?

Depending on your product, usually it makes more sense to go for a small share of a big market if the market is relatively empty of competitors. That way, there’s a much higher potential for growth.

• How will you make money? How much could you make? Give your best estimate.

If you can’t figure out a way to monetize the product, that’s usually a sign that you’re not really solving a big problem. Think about it this way, if someone gave you a cure for aging and you were able to stay forever young, you would probably pay a shit ton of money for it. On the other hand, if someone gave you an app that let you put funny filters on your pets, you probably wouldn’t pay anything for it. If you can monetize your product without drastically shrinking your user base, usually that’s a good sign that you’re solving a real problem. The best way to figure out if people will actually pay for what your making is to go out and ask them. Ex: “Would you pay 20$ a month if I could reduce your time spent on XYZ by 1hr a day?”.

• Vision — if all goes well, what will you have built in 5 years?

Vision is a good indication that the founder is tackling a real problem and is dedicated to solving it. If the product vision is 1–2 years, that’s OK, but it probably won’t have much effect on its users. Imagine if Google had a 2 year timeline, they would’ve never scaled and made the product what it is today.

• Will your market position be defensible 10 and 20 years in the future?

Peter Thiel likes to invest in startups that are solving really challenging problems, like climate change, poverty, etc. He asks this question as a measure of how impactful the product / problem your tackling is. Climate change with 99% still be a problem in 10–20 years. Aging with 99% still be a problem in 10–20 years. Instapaw probably won't, which is why that idea would never get a penny from the Thiel man.

This was an accumulation of some of the stuff I learned over the last few months listening to smart founders and failing myself. Usually if these questions are hard to answer, the product / company couldn’t sustain long term growth.

Resources are finite. The value of a resource is determined by its utility and affluence. It’s something like:

Put it all together...

(New & Old likelihood being the likelihood of achieving XYZ goal) 

We subconsciously make this calculation when we assess the value of a resource. Each resource has a different value based on the individual and the assessment of their condition, but in general we have pretty similar hierarchies of value. 

Here’s a thought experiment:

1. You’re stranded in the desert. You’ve been starving for the last week and a half and are down to your last 100ml of water. A witch comes out of a cave and gives you the choice between a fist full of gold and 1L of water. Which one do you choose?
2. You’re at home, sitting on your couch. A genie comes out of a bottle and gives you the choice between a fist full of gold and a lifetime supply of water. Which do you choose?

Under these assumptions (bellow), we can calculate the value of gold vs water in each scenario:

[likelihood increase scale = 0–1]
[time scale = 0–1]
[resources spent scale = 0–∞]
[quantity available scale = 0–∞]

(affluence is 0–∞ for reasons outlined later. Time is restricted, 1 being all your time, 0 being none. Likelihood is restricted, 1 being 100% 0 being 0%. Resources spent aren't restricted).

In the first scenario:
(Time to acquire resource and physical resources spent are very low [0.1] because it is given to you by the witch, you didn’t have to work hard to acquire it).

In the second scenario:

(Time to acquire resource and physical resources spent are very low [0.1] because it is given to you by the witch, you didn’t have to work hard to acquire it).

Which is why you chose the water in the first scenario, and the gold in the second. We subconsciously pick the items with the highest value.

The equation outlined is pretty basic and doesn’t consider the balance between long term objectives and short term objectives. Short term is prioritized by default over the long term — so if you plugged these into an equation it might look like this:

This is just a guess, I’m not sure how accurate that is (that equation would require the long term value to be >5x that of the short term to prioritize it. It also rises exponentially the more value there is in the short term).

So what happens when there is no limit to quantity?

No limit to amount of resources it takes to acquire?

The net value of all things shrinks to 0 when there is an unlimited amount of resources. It grows to infinity when the resources it takes to acquire the object of value scales infinitely. Because the difference in likelihood is fixed (0-100%), it doesn't impact the final outcome when you have an infinite quantity or an spend infinite resources trying to acquire.

Value relies on limits

Gold is worth $1,481.67 per ounce. Silver is worth $16.92 per ounce. Copper is worth 23 cents per ounce.

Gold isn’t that useful. Silver and copper are much more conductive (if you’re building hardware) and can replace gold at most things. There really isn’t much utility for gold compared to silver, yet its price is 87x higher. Why?

Number of tonnes on the earth:

• Gold: 170K
• Silver: 1.6M
• Copper: 700M

Despites its lack of utility, gold is much rarer than other metals and thus has a much higher price tag. With unlimited tonnes of each, all metals would lose their value.

Another factor is ease of collection. Diamonds for ex are very difficult to extract compared to a common metal like copper. If copper was near impossible to acquire, even though it's abundant, it's value would be very high. 

Recap: we subconsciously create hierarchies of value based on the resources we invested and the affinity of the item. With unlimited resource, there would be no hierarchies of value. Everything would be worth noting. 

Affiliation to action: our actions are thus based off that subconscious value assessment.

In a podcast featuring Vinod Khosla, he talked about how all forms of education (in our current institutions) stem from a handful of core principles. The 1st implication: we don’t need to to teach students everything about everything, we only need to teach them the core principles and they’ll be able to learn the topics of their choice on their own. The 2nd implication: if we train a ML model on these principles, then it could absorb knowledge quickly and teach all students.

I think you could boil all things down to core principles - even something as large as the accumulation of all human behavior. Here are the fundamental laws that I think dictate every one of our actions. 

1. The law of limit and value
2. The law of intrinsic inequality
3. The law of group over man
4. The law of mutual gain
5. The law of sacrificial sustenance
6. The law of extended action ( → effect of truth)
7. The law of extended wealth
8. The law of expected outcome

I’m going to write a post on each.

In September and early October, I spent most of my time building a relationship management tool (RMT). It was supposed to be like linkedin but more personalised and less social (you couldn’t interact with other people). I just wanted a tool that could keep track of the people I’m meeting with, what I’m learning from each meeting. and how I could add value to that person the next time I meet them. This post is an accumulation of the things I learnt over that month and a half. (PS: you can sign up to use the RMT it here).

TL;DR

• Build something you want
• Build features around a key value prop
• Avoid taking shortcuts
• There’s no such thing as cheating
• Sprints help
• Use it

Build something you want

How do you know if an idea is a good one? If it’s something you think you would love to have, build it. An objective metric for evaluating ideas is:

1. How much time is spent using the product
2. How much data is being stored on the product

If users are spending a lot of time using the product (it could be actively using, like reddit, or passively using, like adblocker) but the product isn’t storing large amounts of data, that’s fine. If it’s storing lots of data but isn’t taking up a large portion of the user’s day (like notion), then that’s also fine. If it’s doing none of these, then it’s probably not great.

This applies mostly to SaaS products. I consider companies like AirBNB and Uber in a whole other ball park. Their metrics are based on saving money or time.

Build features around a key value prop

Every feature should boil down to accomplishing one thing. That could be “facilitating the maintenance of relationships and maximizing their mutual value” (for me) or anything that aligns with your vision for the product. If a feature doesn’t align with the main objective, it’s probably not a priority. 

I think the #1 most important thing to build out at the beginning is the functionality. If you complete the functionality and stop using the product after a week, ask yourself why. It might be just not rewarding enough to use because the UI is crappy, in which case you try improving the UI. If you still don’t use it, why? After 10 iterations, you might realise you just don’t like the fundamental principle behind your product. At that point, drop it and do something else.

Avoid taking shortcuts

Building a foundation of rock [in 20hrs] > Building a foundation of sand [in10hrs]

Most of the time, being lazy with a piece of code will bite you in the ass later on when you’re trying to quickly iterate and everything is falling apart. You’ll try and change something, but realize too late you broke XYZ thing in the process, and then try and fix XYZ thing but that breaks this other thing… It’s a big headache that ends up taking you 3x longer to fix than it would have if you spent more time on building good fundamentals.

I say “avoid” instead of “don’t” because there are times where shortcuts make sense. If you’re product seems like a great idea but turns out to be pretty bad after 100hrs of work put in, sure it’s a learning experience but it’s still a waste of time you’ll never get back. Ideally you can build a foundation of rock while executing quickly, but if you can’t and you want to make sure the product is a right fit ASAP, I’d say take the shortcut.

There’s no such thing as cheating

I started building the RMT with raw HTML/CSS/JS. It was a great learning experience for understanding fundamentals, but I think this should be avoided once you get the hang of it. A little later I tried bootstrap studio and felt kind of icky using it cause I could ship an entire website in a couple hours with drag and drop design, but I don’t think that’s the right way to look at tools that drastically reduce shipping time.

Things don’t have to be extremely difficult. We used to program in binary, until we made low level languages and finally high level interpreter languages. It’s easy to put down quick solutions like bootstrap studio because they make things 10x easier, but the reality is we’re already unconsciously living with 10x, even 100x solutions compared to 40 years ago. No one complains about people not using G-Code, or coding neural nets in binary. Optimization is a beautiful thing and should be celebrated not shamed.

I also think at a such an early phase of product design it makes more sense to copy what has already been successful than to completely reinvent the game, especially when it comes to design.

Example #1 — one of the core features of the RMT had to be meeting tracking, and I thought github had a really cool calendar, so I just stole that.

Example #2 — Notion has a sweet sidebar layout for managing your pages, so I just stole that too.

Example #3 — even this “Connections” page layout is strongly based on Notion’s. 

Sprints Help

I realised early on I would need to learn a few things to build an RMT:

1. Javascript (which I realized later when coding everything from scratch in HTML and CSS and couldn’t figure out how to make things interactive without “transform:”)
2. Flask
3. How to manage a database
4. How to host an active page

The most effective thing for learning was trying to build things in 2hr sprints. JS is pretty similar to python in a lot of ways, so it only took a few sprints before I felt confident in making a site interactive. Flask was based in python, so it was pretty easy to pick up. I used firebase for my db, which I could also fiddle with in python. The active hosting part was annoying because I kept on getting errors after uploading and couldn’t figure out why, but google app engine ended up working fine.

The point: if you need to move fast, learn topics that lie on the edge of your circle of competence. Everything I learned was tied to python. You don’t want to jump into a completely unknown territory — you’ll lose motivation quickly.

Use it

Like I said before, the best way to understand how useful your product is is by using it yourself. If you’re not in love with it and don’t use it every day / store a lot of data on it, figure out why and iterate on that. 

I actually still use my RMT, but I don’t have meetings every day so it’s not always useful. I initially thought I’d use it obsessively and always by trying to get back in touch with people who I haven’t talked to in 30+ days, but I just don’t. A big part of that is missing features I wish I could have, but it could also just be the fundamental idea that maintaining relationships with an app just doesn’t have enough payoff.

I don’t really touch it anymore, right now my priorities are elsewhere. But I think the experience of building and launching a product from scratch was really valuable. I hope you learnt something!

1% of the population averages 6.25 hours of sleep a night.

They don’t choose to sleep less — they have a mutation(s) that allows them to naturally wake up at 5am feeling well rested.

A 10 year study lead by Ying-Hui Fu and Louis Ptáček uncovered the cellular level processes that lead to the condition. There are three known mutations.

DEC2

MyoD1 is a gene that encodes a protein that promotes orexin production. Orexin is a neuropeptide that regulates arousal, wakefulness, and appetite. DEC2 is a protein that binds to MyoD1 to inhibit transcription by RNA polymerase. 

When DEC2 functions properly, orexin inhibition is in sync with the circadian rhythm. As it gets dark outside, your brain produces melatonin in the pineal gland, which basically inhibits certain parts of the brain when you fall asleep.

Flux of DEC2 production is very similar to that of melatonin’s. During the day, production is low. During the night, production is high. 

But for the 1% of people that have the mutation, their DEC2 proteins don’t bind very strongly to MyoD1, meaning RNA polymerase can still synthesize the protein that increases orexin production. This effects orexin production closer to dawn, when DEC2 production starts to decrease.

ADRB1

The ADRB1 gene encodes the protein beta-1 adrenergic receptor. Beta-1 adrenergic receptors act as neuron inhibitors — when they receive agonist chemical signals, they inhibit major processes in the neuron. When ADAB1 is mutated, beta-1 adrenergic receptor degrades much faster. As they degrade, they lose their ability to inhibit function, meaning neurons function longer, and a person can stay awake longer.

NPSR1

They had discovered DEC2 and ADRB1 mutations in most patients with the condition, but couldn’t find it in some. So they took a family that all had the condition but didn’t have DEC2 or ADRB1 mutations and looked at their dorsal pons brain tissue. They found a mutation in NPSR1.

NPSR1 encodes a G protein-coupled receptor that binds to neuropeptide S (NPS). NPS is mainly produced in the amygdala and spreads within the Thalamus region. It induces wakefulness and arousal. 

Those with the mutation have more sensitive receptors, meaning they feel more awake with low levels of NPS. They inflicted mice with the mutation to validate that it had something to do with sleep. They found that those mice slept 59min less on average (52min less deep sleep and 7min less REM sleep) while still waking up rested.

Effect

Sleep deprivation effects cognitive performance and health poorly.

1) It reduces cognitive abilities. There are lots of studies showing decrease in performance the longer the cumulative deprivation. Here’s a graph taken from this paper. (Black squares = 0hrs, Circles = 4hrs, White squares = 6hrs, Diamonds = 8hrs).

• PVT lapses → lapses in psychomotor vigilance task (basically the amount of times you stop focusing on something that you’re supposed to, like “follow this dot as it moves around the screen”).
• DSST → digit symbol substitution task (basically substituting symbols in a digital graph)

2) Increases mortality rate by 13% (if you want to look more into this, I made a paper here).

You would expect people with these mutations to have terrible cognitive abilities and be riddled with health conditions, but they’re not.

There isn’t as much research in this space, but Ying-Hui Fu and Louis Ptáček claim these mutations don’t cause any significant health defects. Some studies even show they increase life span (although sample sizes are small).

Implication

If these mutations are only beneficial, why wouldn’t we all have them? The leading theory is that it they only came up a few thousand years ago. Because of technological progression, it hasn’t had large effects on evolution (we aren’t struggling for food, we aren’t struggling to survive, reduction of sleep doesn’t effect us as much as it would have 200K years ago).

Inducing a mutation is very invasive. It’s also pretty much impossible to do in all neurons in the brain of a full grown adult. Alternatively, one of the most popular non invasive ways to influence neuron activity is transcranial stimulation.

If we could reduce the inhibitory effects of DEC2, reduce the level of inhibitive capabilities of beta-1 adrenergic receptors (involved in ADRB1), or increase the sensitivity of NPSR1 G protein encoded receptors to NPS, we could reduce sleep time without harming cognitive processes or long term health.

When I was younger, I was a trouble maker. 

I think it came from prioritising my happiness above other peoples’ and being curious to see what the outcomes of my actions would be. For ex, in kindergarten my parents would meet with my teacher every day after school because I started a fight, wasn’t raising my hand to talk, didn’t focus on activities, etc. 

I was too focused on being happy right now to think about the outcomes of my actions. After the damage was done, I would feel horrible but forget about that feeling and do it again the next day. My mom figured out a way to make me reflect on my actions — she would ask:

“What are the 3 things you want people to say about you at your funeral? Do you think they would say those things about you if you died today?”

I wanted people to say that I was:

1. Kind
2. Driven
3. Fun to be around

And none of those reflected my actions, so I started to change. I started to pause before doing mischievous things to think about the outcome on peoples’ perception of me. It wasn’t long till I changed my habits completely.

Today I don’t think there are only 3 things I want people to say about me when I die — I think there are a lot more. I made a list of all the things I think make an ideal person.

Which can be generalised as:

1. Has emotional control
2. Isn’t a slave to external factors 
3. Understands themselves and the cause of their actions
4. Cares deeply for themselves and for others
5. Is ambitious
6. Is optimistic

Understanding what I want to become -> understanding how I'll become.

Your character is the result of an accumulation of experiences. Your experiences are interpreted by your desires. A positive experience is the outcome of experience aligning with desire. A negative one is the outcome of experience not aligning with desire.

Experience is easier to control than desire. Desire arises naturally as an outcome of emotion, but experience is usually voluntary. 

So, understanding what you desire is the best way to understand what your experience should be. Right now, I’m at a crossroads. Should I go to XYZ university, take a gap year, work at a startup, get a PhD, etc? There is a seemingly infinite amount of choices I could make, but only a finite amount of choices align with my desire of an ideal person — I just need to figure out what those are.

Hierarchy of Desire

Desire shapes the interpretation of experience, but is also the outcome of experience. Your emotional response to an experience dictates how much you desire it.

As you experience more, desire reaches a local minimum where you understand what your priorities are relative to your environment. If you could experience everything there is to experience, you could theoretically reach a global minimum. 

Experience passed on through many millennia of generations (through stories and genetic influence) have conditioned us to desire things hierarchically. 

1. The first block is primary needs. If someone does a “bad” thing, like stealing or putting another person down, it’s because their primary stack isn’t fulfilled. 11% of Canadians struggle with food insecurity and can’t even hit the first block.
2. You can only feel a high sense of esteem once you feel a sense of belonging to a community. It makes sense —esteem is the feeling that you’re valued, and value comes from the benefit your actions have on other people, which can only be nurtured when you’re immersed in a trustworthy community.
3. Seeking understanding and beauty (in nature, in art, etc) are forms of value extraction.
4. Self actualisation and transcendent needs are hand in hand. Transcendence is feeling personally fulfilled and wanting to now focus on helping other people. Self actualisation (feeling like you’re making the right decisions with the right intentions) relies heavily on how much value you’re adding to other people. If other people aren’t benefiting from your actions, you probably won’t feel self actualised.

Lower level needs tend to give quick and intense feelings of happiness. They are priority #1 so our bodies have evolved to make us feel especially happy when we accomplish them, but only for a short time. When focusing on primary needs, we have a large margin of potential happiness and a low general state.

Higher level needs increase our happiness state, not just our current level. They also lower the margin for positive and negative emotion, making us emotionally stable.

The traits I outlined for what I think is an ideal person fall into the hierarchy like this:

1. ✖ Physiological
2. ✖ Safety
3. ✖ Belonging
4. ✔ Esteem : (2) Isn’t a slave to external factors, (4) Cares deeply for themselves
5. ✔ Understanding + Beauty : (1) Has emotional control, (3) Understands themselves and the cause of their actions
6. ✔ Self Actualisation + Transcendent need : (4) Cares deeply for others, (5) Is ambitious, (6) Is optimistic

I'm lucky enough to be born in Canada. The first 3 blocks have pretty much always been satisfied.

I'll base my next long term experience off how it aligns the last three, not off how it makes me or feel in the moment.

If you don’t know who you are and who you want to be, you’ll become what other people want you to be. Your desire will be out shined by other peoples’ desire for you. Use a top down framework to figure out what you should do. Let me know if it works.

We wanted to take this concept further, and visualize what the human eye sees from EEG data. You can read the paper we published here, but that’s not the point. I noticed that all the projects we were working on were time crunched and were never replications of other neuroscience papers.

Our lab operated on “90 minute sprints”, a method inspired by MIT’s lab (Our boss was a professor at MIT before this).

If any results came out of a project before 90 min, we would keep working on it in the next sprint. If there were no results, we would abandon the project and move to something else.

This high intensity work flow was great for developing deep CS + hardware understanding, but there was always a voice in the back of my head that would question how valuable they were.

Microsoft wasn’t built in 90 minutes, Tesla wasn’t built in 90 minutes, PayPal wasn’t built in 90 minutes, CRISPR wasn’t discovered in 90 minutes. We were trying to 1) Make breakthroughs 2) Make them quickly, which seemed impossible.

So I started looking into the incentive structure for research facilities and the outcome on the quality of research generated. 

Academic incentives are messed up.

What I found was that research facilities across North-America have an incentive structure that doesn't promote good quality research.

Scientific Status

• Novelty and constant ground-breaking progress brings higher scientific status than long term research or exhaustive studies.
• Short term projects bearing little fruit aren't as risky as long term projects bearing little fruit.

Replication

• Replication + validation of previous work is discouraged.
• Independent, direct replications of others’ findings can be time-consuming for the replicating researcher.
• Replications are likely to take energy and resources away from other projects that reflect one’s own original thinking.
• Replications are generally harder to publish (because they’re viewed as being unoriginal).
• Even if replications are published, they are likely seen as ‘bricklaying’ exercises, rather than as major contributions to the field.

Validation

• Hypothesis validation brings more social clout than hypothesis invalidation. Ex: imagine spending 30 years on a long term study of cancer to realise your hypothesis for a cure isn’t viable. This is a big fear for researchers.
• Research that isn’t successful at supporting a hypothesis gets kept away in the file cabinet, which causes a lack of transparency in the space.

Criticism

• Criticism of the current academic incentive system only hurts your reputation.
• Researchers who stand up for true science and complain about “fake” research are perceived as being incompetent enough to succeed in the current hierarchy. They are seen as complainers who aren’t well suited enough for proper scientific study.
• Ex: “Princeton University psychologist Susan Fiske drew controversy for calling out critics of psychology. She called these unnamed “adversaries” names such as “methodological terrorist” and “self-appointed data police”, and said that criticism of psychology should only be expressed in private or through contacting the journals. Columbia University statistician and political scientist Andrew Gelman, “well-respected among the researchers driving the replication debate”, responded to Fiske, saying that she had found herself willing to tolerate the “dead paradigm” of faulty statistics and had refused to retract publications even when errors were pointed out. He added that her tenure as editor has been abysmal and that a number of published papers edited by her were found to be based on extremely weak statistics; one of Fiske’s own published papers had a major statistical error and “impossible” conclusions.”

As a result, the current research space has become saturated with short term projects with inflated results.

The number of faculty positions in the research space has remained in proportion with population growth, but cumulative PhDs awarded has not remained in proportion to faculty positions. 

More PhDs are being handed out then ever before. This could either mean that 1) Researchers are getting much better at picking interesting topics to pursue and are finding much more concrete data supporting their thesis’s (could be due to technological advancement). 2) Researchers are inflating results to get their hands on PhDs. 

The first could be true in an economy that’s rapidly growing with tech. The more competent the tech, the more data could be gathered supporting a hypothesis. Plus, the more economically incentivized research would become due to it’s high demand, which could explain the significant rise in PhDs.

GDP is the best measure of economic growth.

But growth is shrinking over time, which disproves the first hypothesis and further proves the second. 

Troubling Stats

More than 50% of researchers have use >1 questionable work practice (QRPs)

Examples of QRPs

• selective reporting or partial publication of data (reporting only some of the study conditions or collected dependent measures in a publication)
• optional stopping (choosing when to stop data collection, often based on statistical significance of tests)
• p-value rounding (rounding p-values down to 0.05 to suggest statistical significance)
• file drawer effect (nonpublication of data)
• post-hoc storytelling (framing exploratory analyses as confirmatory analyses)
• manipulation of outliers (either removing outliers or leaving outliers in a dataset to cause a statistical test to be significant).

This has had major cumulative effects on the ability to replicate other papers:

“According to a 2016 poll of 1,500 scientists reported that 70% of them had failed to reproduce at least one other scientist’s experiment (50% had failed to reproduce one of their own experiments).”

It’s especially present in fields that fall subject to pseudo-science, like psychology and neuroscience (I remember altering some of the sample data I was gathering to make it fit my hypothesis back at the lab. I only did it for one test, but it was definitely tempting later on).

Psychology

Replicating trials in 2012 were done to see if older psychology papers could be replicated.

• 91% of new papers with author overlap from previous papers could be replicated
• 64% of those without couldn’t

Another set of replicating trials in 2015 examined the reprehensibility of the top 100 psychology papers from top science journals.

• 36% had significant findings compared to the 97% stated in the journal

When you look deeper into psychology, different sub-fields have different replication results.

• Cognitive psychology replication rate = 50%
• Social psychology replication rate = 25%

Papers with multiple researchers on board yield better replication results

• 91.7% with multiple researchers on board
• 64.6% without

Basically, older authors have more legitimacy and have a reputation to preserve, newer authors don’t. Plus, fields that rely on low amounts concrete data to make assumptions are probably not replicatable. 

When there are multiple researchers on board, they keep each other accountable and their work yields more accurate results.

Why is it happening now?

Like mentioned before, the incentive structure for researchers doesn’t encourage paper replication and exhaustive or long term studies. The current “Publish or Perish” mentality is synchronised with the “Move fast and break things” mentality of startups.

The field is saturated with ideas and hypothesis’, but no one’s willing to put in the long hours to validate or invalidate them, so shortcuts are taken (inflated results and false positives). 

This wasn't a problem 50 years ago. It only became a problem because:

1. Big companies have been progressively substituting expensive and timely in-house research for university contracted research, to even cheaper and faster 3rd party research companies. Problem: companies want results reinforcing the use of their product. 1) creates bias towards validating hypothesis and not disproving them. 2) research companies in competition fight to bring companies the best and fastest results, cutting corners in the process.
2. As the scientific community becomes more dense, it becomes harder to stand out. Without groundbreaking results, it’s very difficult to seem legitimate and competent in your field.
3. There is an increasing stress on fast paced development. There are much fewer long term studies being executed on (relative to scientific community size) than there were in the 1950s. Global access and competition accelerates development but also squashes patience and validity.
   
   

What we can do about it

There doesn’t seem to be one solution. But, if you break down how a research paper is made, you can identify key areas where researchers throw in their biases.

Here are a few ways to tackle the problem:

• Tackling publication bias with pre-registration of studies→pre-determine and peer review your method for data collection / validation. Incentivises follow through and publication in the case of results contradicting hypothesis, and outcome alteration (QRPs)
• Emphasise replication attempts in teaching
• Reduce the p-value required for claiming significance of new results
• Address the misinterpretation of p-values
• Encourage larger sample sizes
• Share raw data in online repositories
• Fund replication studies specifically
• Emphasise triangulation, not just replication→attempt other methods challenging the problem and weigh different outcomes for each

I see pre-registration of studies and sharing data online in large repositories as some of the best ways to face the problem right now. I'm not sure how we'll get around it, but faulty research is laying down a foundation of sand and will cause huge long term problems if we don't find a way to restructure incentives.

If you want to learn more, check out this podcast with Peter Thiel and Eric Weinstein.